Introduction Decentralisation of healthcare commissioning in the UK has led to the development of multiple models of community-based care for COPD. Despite the huge costs associated with these long term condition (LTC) teams there is little published data exploring the efficacy of these models of care. Until 2009 healthcare in Gwent, South Wales was delivered by five autonomous Local Health Boards (LHB). Different models of community-based COPD care were developed in each. This gave a unique opportunity to compare the efficacy of different models of healthcare among patients with similar demographic characteristics. The three main models of care were similar in primary objective: to reduce hospital admissions. In Caerphilly LHB COPD was managed by nurse specialists. In Newport and Monmouthshire generic nurses managed COPD with other chronic conditions. LTC nurses in these LHBs worked alongside GPs without secondary care. Services were introduced to some practices but not others. In Torfaen and in Blaenau Gwent COPD was managed by nurse specialists, GPs and Consultant Chest Physicians.

Methods Hospital admission data were collected from the trust database. COPD admission data were compared for the 3 years before and after the introduction of services around 2006.

Results Gwentwide 2003–2009 there was a 5.5% increase in COPD admissions. In Caerphilly the increase was 12.3%. In Newport and Monmouthshire for those practices working with LTC there was an increase of 20.8% and 43.4% respectively. This corresponded to an increase of 5.2% and 1.0% respectively in those practices in Newport and Monmouthshire that had no contact with LTC teams and continued to manage COPD in the same way (p<0.01 for within LHB comparison) In Blaenau Gwent and Torfaen, however, the inclusion of secondary care was associated with a reduction in admissions of 3.8% and 5.1% respectively. (p<0.01 for the comparison between pooled data for Torfaen/Blaenau Gwent and Newport/Monmouthshire/Caerphilly).

Conclusion There is little evidence base to guide the huge sums of money invested in community care for COPD and other chronic conditions. Our data suggests that services with close communication between primary and secondary care may be more effective than those developed around primary care alone.